THE CRISIS IN PATENT COVERAGE: DEFINING SCOPE OF AN...

V o l u m e 8, N u m b e r 1 Fall 1994

THE CRISIS IN PATENT COVERAGE: DEFINING SCOPE OF AN INVENTION BY

FUNCTION

P a u l M. J a n i c k e *

INTRODUCTION

One o f the mos t impor tan t issues in the cu r ren t debate about the role

o f intel lectual p roper ty r ights in m o d e r n wor ld t rade is the quest ion o f

patent scope. G i v e n that a pa ten t might be obta inable for an invent ion,

how b road should that pa ten t be? How much should it cover , and thereby

prec lude o thers f rom using? T he p r ob l em has been the subject o f genera l

address by c o m m e n t a t o r s ; ~ o f analysis wi th respect to par t icular indus-

t r ies ; : and of changes in recen t years to the Uni ted States patent laws

re la t ing to scope o f pa ten t r ights . 3

* Professor of Patent Law, University of Houston Law Center. B.E.E., Manhattan College; J.D., New York University; LL.M. (Patent and Trade Regulation Law), George Washington'University. Of counsel, Arnold, White & Durkee.

1. See, e.g., Robert P. Merges & Richard R. Nelson, On the Complex Economics of Patent Scope, 90 COLUM. L. REV. 839, 843 (1990) (excellent review of many of the problems of patent scope, arguing that the effects of broad patent scope depend upon (i) the relationship between successive technical advances in the particular industry, and (ii) the extent to which firms license technologies to each other); Rafael X. Zahralddin, The Effect of Broad Patent Scope on the Competitiveness of United States Industry, 17 DEL. J. CORP. L. 949 (1992) (arguing that unduly broad patents are an impediment to economic progress in some industries, notably semiconductors and software).

2. See, e.g., Pamela Samuelson, Benson Revisited: The Case Against Patent Protection for Algorithms and Other Computer Program-Related Inventions, 39 EMORY L.J. 1025 (1990) (critiquing lower courts' broad interpretation of Diamond v. Diehr, 450 U.S. 175 (1981), which permitted patenting of some computer-implementable processes); see Edward T. Lentz, Adequacy of Disclosures of Biotechnology Inventions, 16 AMER. INTELL. PROP. L.A.Q.J. 314, 324 (1988-1989) (commenting on seemingly excessive breadth of biotechnology patents vis-b.-vis the technical disclosures in these patents); Gary Slutsker & David C. Churbuck, Whose Invention Is It Anyway? FORBES, Aug. 19, 1991, at 114 (criticizing the patent system with regard to biotechnology inventions).

3. For example, the section of the patent statute that defines acts of patent infringement is 35 U.S.C. § 271. That section was narrowed in 1984 to provide an infringement exemption for making, using, or selling a patented invention "solely for uses reasonably related to the development and submission of information" to the Food and Drug Administration, typically in seeking FDA approval of a drug or medical device. 35 U.S.C. § 271(e), added by Pub. L. No. 98-417, § 202, 98 Stat. 1603. Section 271 was broadened in 1988 to make unlawful, inter alia, the importation of an unpatented product made by a process carded out abroad, where the process, if practiced in the United States, would have violated a U.S. process patent. See 35 U.S.C. § 271(g) (1988), added by Pub. L. No. I00- 418, § 9003, 102 Stat. 1563-64.

156 Harvard Journal of Law & Technology [Vol. 8

This article will address one controversial aspect of patent scope, 4 the

use of functional language in patent claims. This aspect arises when one

defines an invention by what its parts do rather than by the structural

names of the parts. 5 This paper will therefore analyze the development

of what may be styled "means-plus-function law" as it applies to several

distinct questions of patent interpretation. 6

The current patent statute, enacted in 1952, 7 was the first to permit

expression of one or more elements of a combination patent claim in

terms of "means for" accomplishing a stated function, s Attorneys

4. Beyond the scope of this article are other currently important aspects of patent scope, such as applying the doctrine of equivalents to expand patent coverage. For a discussion of that topic, see, e.g., Martin J. Adelman & Gary L. Francione, The Doctrine of Equivalents hi Patent Law: Questions That Pennwalt Did Not Answer, 137 U. PA. L. RI~V. 673 (1989). The main issues affecting the doctrine of equivalents are, at the time of this writing, before the U.S. Court of Appeals for the Federal Circuit for en banc review, in Hilton Davis Chem. Co. v. Warner-Jenkinson Co., Appeal No. 93-1088, 1993 U.S. App. LEXIS 38138. See also En Banc Federal Circuit Will Review Doctrine of Equivalents and Equity, 47 Pat. Trademark & Copyright J. (BNA) 133 (1993). Also beyond the scope of this article is the question of whether a patent on a biological material should be allowed to block all uses of that material during the life of the patent. On this subject, see Thomas D. Kiley, Patents On Random Complementary DNA Fragments? 257 SCIENCE 915 (1992) (arguing that disclosure of a trivial utility should not allow a patentee to block more important uses discovered later for biologic material); P. Kelly, ProlJhetic Patents in Biotechnology, 8 BIo/TEclI. 24, 25 (1990) (criticizing use of imaginary, non-laboratory-backed examples in patent specifications on biology subjects). Finally, the present work does not cover whether experimental use is, or should be, generally exempt from the reach of a patent claim for infringement purposes. See, e.g., Rebecca S. Eisenberg, Patents and the Progress of Science: Exclusive Rights and Experimental Use, 56 U. Cltt. L. REV. 1017, 1070-74 (1989) (advocating greater employment of the experimental use defense against patent infringement); Irving N. Felt, Biotechnology Research and the Experimental Use Exception to Patent Infringement, 71 J. PAT. & TRADEMARK OFF. SOC'Y 819 (1989) (arguing for compulsory licenses to support experimentation by others with patented materials).

5. For earlier works on this particular aspect of patent scope, see 4 Donald S. Chisum, Patents (MB) § 18.0315l (Nov. 1994); R. Carl Moy, The Interpretation of Means Expressions During Prosecution, 68 J. PAT. & TRADEMARK OFF. SOC'Y 246 (1986); Maxim H. Waldbaum & David Sipiora, Pennwalt Redux--Judicial Uncertainty vs. Procrustean Bed, 19 AMER. INTELL. PROP. L.A.Q.J. 237 (1991); Kenneth R. Adamo, The Double Standard--bz re Bond, The Office, 35 U.S.C. § ]12, para. 6 anti Atlantic Thermoplastics Co., 2 FED. CtR. B.J. 137 (1992); Donald S. Chisum, Patent Law Developments in the United States Court of Appeals for the Federal Circuit During 1991, 41 AM. U. L. REV. 869, 903-07 (1992). See also this author's address to a conference sponsored by the American Law Institute and American Bar Association on Patent Law Developments, Washington, D.C., December 5-6, 1992, from which the present article in part evolved: Paul M. Janicke, Means-Plus-Function Claims in Modern Patent Law, Dec. 4, 1992, available in WESTLAW, ALI-ABA database, C785 ALI-ABA 297.

6. These include (i) defining novelty over the prior art; (ii) defining nonobvious subject matter; (iii) defining eligible subject matter for patenting; and (iv) defining the breadth of patents for infringement purposes.

7. Ch. 950, 66 Stat. 792 (July 19, 1952) (codified as amended at 35 U.S.C. § 1 et seq. (1988))

8. See 35 U.S.C.A. § 112, para. 6 (West Supp. 1992); infra note 13. The newness of this

No. 1] Crisis in Patent Coverage 157

drafting patent applications were quick to adopt that style of claim

language, The U.S, Court of Appeals for the Federal Circuit 9 over the

last decade has taken a close look at the implications of using means-plus-

function terminology in claims, culminating in two 1994 en banc I°

decisions on the subject: In re Dona ldson Co, jl and In re Alappat , 12

The Patent and Trademark Office ("PTO") for a time refused to

follow the court in its decisions on the subject, leading to considerable

strain between the court and the PTO, That strain has been largely or

fully resolved by Donaldson and Alappat , and the law now appears finally

to have come full circle to where it stood before 1946, when a now-

famous Supreme Court decision started the conflict,

I. THE PATENT STATUTE PERMITS

FUNCTIONAL DEFINITIONS

The statutory provision for functional patent claim language is in the

sixth paragraph of 35 U.S.C. Section 112:

An element in a claim for a combination may be expressed

as a means or step for performing a specified function

without the recital of structure, material, or acts in support

thereof, and such claim shall be construed to cover the

corresponding structure, material, or acts described in the

specification and equivalents thereof.13

This provision functions mainly through two parts. The first says that

one can express an element of a claimed combinat ion by reciting the

words "means for" followed by the function that element performs in the

provision is described in P.J. Federico, Commentary On the New Patent Act, 35 U.S.C.A. 1, 25-26 (West 1954).

9. The Court of Appeals for the Federal Circuit was established under the Federal Courts Improvement Act of 1982, Pub. L. No. 97-164, § 101, 96 Stat. 25 (1982), and has exclusive appellate jurisdiction of all appeals from district courts in patent cases.

10. The enabling statute, 28 U.S.C. § 46(c), refers to a proceeding before all the judges of the court as an "in banc ~ proceeding. However, for consistency, this paper will refer to all such proceedings as being "en banc."

11. 16 F.3d 1189 (Fed. Cir. 1994). See discussion supra accompanying notes 122-33. 12. 33 F.3d 1526 (Fed. Cir. 1994). See discussion supra accompanying notes 84-108. 13. 35 U.S.C.A. § 112, para. 6 (West Supp. 1992). Although the paragraphing has been

changed since 1952 (this provision was originally the third paragraph), the text of the provision has not been amended in any way.

158 Harvar d Journal o f L a w & Technology [Vol . 8

c l a i m e d c o m b i n a t i o n . ~4 Pa t en t c l a ims con t a in ing such l anguage a re

n u m e r o u s . 15 T h e conf l i c t has c e n t e r e d a r o u n d the s e c o n d par t o f the

p rov i s ion ; " a n d such c l a i m shall be c o n s t r u e d to c o v e r the c o r r e s p o n d i n g

s t ruc tu re , ma te r i a l , o r ac ts d e s c r i b e d in the spec i f i c a t i on and equ iva l en t s

the reof . ''t~ At f i rs t it m a y a p p e a r that this p r o v i s i o n g ives func t iona l

r ec i t a t ions a ve ry b r o a d c o v e r a g e indeed , p e r h a p s r e a c h i n g any and all

s t ruc tu re s n o w k n o w n o r la ter d e v e l o p e d w h i c h can c a r r y out the rec i t ed

func t ion . R e c e n t ca se law s h o w s that such a v i ew w o u l d be w r o n g , and

that 1 12(6) ac tua l ly a c c o r d s far m o r e r e s t r i c t ed c o v e r a g e . Th i s " cu t t i ng

b a c k " e f f e c t o f 1 12(6) w a s f irs t e x p r e s s l y o b s e r v e d by the Fede ra l Ci rcu i t

14. The statute '.dso says that an element can be recited as a "step" fur achieving a function. Id. This would appear to apply to language in process claims, i.e., where a combination of process steps constitutes the claimed subject matter. Moreover, the words following "without the recital of . . . acts in support thereof" would lead to the same conclusion. At the time of enactment of the statute there was authority condemning method claims for functionality, i.e., for being too result-oriented. See Universal Oil Prods. Co. v. Globe Oil & Refining Co., 322 U.S. 471,484 (1944). Present case law is sparse on just what kinds of steps are functional. See hi re Angstadt, 537 F.2d 498 (C.C.P.A. 1976) (recognizing that functional method claim language is sanctioned by the last paragraph of Section 112); bz re Roberts. 470 F.2d 1399, 1402-03 (C.C.P.A. 1973) (permitting a method claim to recite the step of "reducing the coefficient of friction o f . . . film to below about 0.40 as determined by [a conventionall test"); h~ re Cohn, 438 F.2d 989, 991 (C.C.P.A. 1971) (method step can, pursuant to last paragraph of Section 112, be recited as doing something "until" .'t particular result occurs, provided the result can in practice be determined with adequate precision by persons skilled in the art). However, no case has been found interpreting this "or step" aspect of paragraph 6 as it appears in the patent statute, and there is no published writing explaining the intent of its drafters. It remains to be seen under what circumstances a process step will be regarded as impacted by the narrowing effect of 112(6). Perhaps it will be those steps which are result-oriented, e.g., heating said solution to deposit metallic iron.

15. In 1993 alone there were 42,082 U.S. patents issued with the expression "means for" in their claims. Search of LEXIS, Patent library, Util file (Oct. 22, 1994).

16. 35 U.S.C.A. § 112, para. 6. Care must be taken to distinguish this statutory provision from the equitable doctrine of equivalents. See it~¢ra notes 133, 149-63 and accompanying text. Briefly, the equitable doctrine operates to expand patent coverage beyond the language of the claims, where equity requires such expansion to prevent a "fraud on the patent." Graver Tank & Mfg. Co. v. Linde Air Prods. Co., 339 U.S. 605, 608 (1950). To accomplish this, a claim is read as covering not only all the different embodiments of technology expressed in the claim's words (which are typically broad and generic to ~tart with), but also any other embodiments which operate in "substantially the same way and accomplish substantially the same result" as those within the claim. Id.

112(6) operates in a similar way, but from a radic~xlly different starting point. In most technologies the disclosure by a patent applicant of a single embodiment in her patent specification will support a claim of generic breadth, limited only in that it must not cover any embodiment known in prior literature or products. Hence many applicants disclose in the specification only one or two embodiments covered by their generic claim. The central point of this article is: 112(6) says that where a functional expression is used, a determination of coverage is made by starting not with the generic claim language but with the often narrow structures shown in the disclosure portion of the patent. Equivalent structures are then added in tn complete the coverage analysis.

No. 11 Crisis in Patent Coverage 159

in 1987 j7 and has been reinforced since then. j8 It turns out that the

coverage now accorded to such expressions is exactly what was sought by

a certain inventor in the 1930s and refused by the Supreme Court in

Halliburton Oil Well Cementing Co. v. Walker• 19

To frame the background for Halliburton and for the controversy that

ensued from it, a few basic principles of patent law must be reviewed.

The first is that a patent application has two main components of interest

here: (i) the disclosure portion, which includes the specification and

drawings; and (ii) the claims. ~-° Since disclosure of a single embodiment

or version of the invention normally will support broad, generic claims,"

and since the claims are what traditionally have been thought to define a

pa!ent's coverage, "~-~ the disclosure portions of most patents have been

narrowly drafted to concentrate on one or two preferred embodiments•

The claims, by contrast, are broadly worded in order to secure as much

coverage for the patent as possible. When courts later interpret the claims

in infringement actions, they generally do so without limiting the claims

17. See Data Line Corp. v. Micro Technologies, Inc., 813 F.2d 1196, 1201 (Fed. Cir. 1987) (112(6) acts as a restriction on the literal satisfaction of a claim limitation); Pennwalt Corp. v. Durand-Wayland, Inc., 833 F.2d 931,934 (Fed. Cir. 1987). cert. denied, 485 U.S. 961, cert. denied, 485 U.S. 1009 (1988). See also Texas Instruments, Inc. v. United States Int'l Trade Comm'n, 805 F.2d 1558 (Fed. Cir. 1986). While Texas htstruments did not mention the cutting-back effect of 112(6), the case probably represents the most dramatic example in the reported literature of that effect in action. The court affirmed an ITC administrative law judge's holding of noninfringement, even though the judge had found, and the appellate court agreed, that "every function in the [Texas Instruments] patent claims is performed by the accused calculators." ld. at 1567. The court's reasoning will be examined hafra, part V.A.

18. See Johnston v. IVAC Corp., 885 F.2d 1574, 1580 (Fed. Cir. 1989) (paragraph 6 "operates to cut back on the type of means which could literally satisfy the claim language"); Jonsson v. Stanley Works. 903 F.2d 812, 819 (Fed. Cir. 1990) (same, citing Johnston).

19. 329 I I.S. 1 (1946). 20. See 35 U.S.C. § 112, paras. 1-2 (1975). This section actually refers to the claims as

the concluding part of the specification, but it is customary practice to speak of the claims as being apart from the specification because of the paramount importance of claim breadth in securing patent coverage.

21. The specification only needs to disclose one way of making and using the generically claimed subject matter, See 35 U.S.C. § 112, para. 1 (1975) (specification to set forth best mode contemplated by inventor for carrying out his invention); 37 C.F.R. § 1.71(b) (1988) (specification "must describe completely a specific embodiment of" the claimed invention). For this reason it was customary to draft the detailed portion of the specification narrowly. See the "preferred layout" of an application as recommended in U.S. Dep't of Commerce, Manual of Patent Examining Procedure § 608.01 (Mar. 1994).

22. See, e.g., Robert L. Harmon, PATENTS AND TttE FEDERAL CIRcurr 13 (3d ed. 1994) (claim provides the metes and bounds of the right which the patent confers on the patentee to exclude others from making, using, or selling the protected invention). The claim is not interpreted as being limited to the preferred embodiments or specific examples contained in the specification. Id.

160 Harvard Journal of Law & Technology I Vol. 8

to the particular structures making up the embodiments shown in the

patent's disclosure. 23

II. THE ROOT OF THE PROBLEM:

HALLIBURTON v. WALKER

The origins of 112(6) lie in the split of authority before the 1952 Act

as to whether means-plus-function recitations in patent claims were (i)

inherently indefinite, 24 (ii) indefinite only in claims drawn to a single

means, z5 (iii) indefinite only if used at the "point o f novelty, "~* or (iv) not

indefinite at all. '7 In particular, the drafters were attempting to address

the 1946 Supreme Court decision in Halliburton Oil Well Cementing Co.

v. Walker, '8 which held such language to be fatali!/indefinite. An under-

standing of the details of that case is essential to an understanding of the

full circle in which the law has moved between 1946 and 1994 with

respect to means-plus-function definitions of patented inventions.

23. See, e.g., Texas Instruments, Inc. v. United States lnt'l Trade Comm'n, 805 F.2d 1558, 1562 (Fed. Cir. 1986) (all possible variations are not required to be set out in the specification, and the court has cautioned against limiting the claimed invention to preferred embodiments or specific examples in the specification); Sjolund v. Musland, 847 F.2d 1573, 1582 (Fed. Cir. 1988) (general rule is that limitations from the specification are not to be read into the claims, since the claims define the invention).

24, See, e.g., Tyden v. Ohio Table Co., 152 F. 183, 185 (6th Cir. 1907) (claim reciting means-plus-function is "substantially a claim for a function" which "cannot be upheld and must be held void").

25. See General Elec. Co. v. Wabash Appliance Corp., 304 U.S. 364, 368 (1938) (claim to lamp filament, in functional terms, was indefinite).

26. See, e.g., Mead Johnson & Co. v. Hillman's, Inc., 135 F.2d 955, 958 (7th Cir.) ("functional statements cannot be availed of to define over the prior art"), cert. denied, 320 U.S. 752 (1943); Farmers' Coop. Exch. v. Turnbow, 111 F.2d 728, 732 (9th Cir.) (claim invalid for using "conveniently functional language at the exact point of novelty"), cert. denied, 311 U.S. 681 (1940).

27. See, e.g., Continental Paper Bag Co. v. Eastern Paper Bag Co., 210 U.S. 405 (1908) (claim upheld as valid, reciting "means for operating the forming plate" and ~means to cause said |~:,:qze to oscillate ~ in combination with other elements).

28,. 329 U.S. I (1946). For the proposition that 112(6) was intended to address Halliburton, see Charles J. Zinn, COMMENTARY ON NEW TITLE 35, U.S. CODE ~PAT- E N T S , ~ reprinted in 1952 U.S.C.C.A.N. 2507, 2514. Mr. Zinn was law revision counsel to the House Judiciary Committee worl:.ing on the bill, H.R. 7794, which became the 1952 Act. See also In re Donaldson Co., 16 F.3d 1189, 1194 (Fed Cir. 1994) (112(6) enacted "to statutofily overrule" holding in Halliburton).


A. The Walker Depth-Measuring Scheme

Mr. Walker was the plaintiff-patentee in the case, and the Hall iburton

company was the accused infringer. Walker was an oil field consultant.

The Hall iburton company was a large oil well services company that met

the needs of members of the oil industry. The company cemented walls

of wells to prevent earth cave-ins, perforated rock formations at the

bottom of wells to permit greater oil flow, and assisted in pumping oil out

of wells.

The purpose of Walker ' s patented invention was to make sure that

during production of oil the downhole pump was always located

substantially below the oil surface, hence in the proper position to draw

oil out. -~9 Prior efforts to do this had injected a sound pulse down into the

well from the ground, and then "listened" at the ground to determine how

long it took for the pulse echoes to travel down to the surface of the oil

pool and back to ground level again. This method led to erroneous

results. 3°

Mr. Walker , like others before him, realized that the error inherent in

prior measurements was due to the unpredictabili ty of the velocity of

sound in oil wells. He devised a system which bypassed that problem.

His invention "l is tened" not only for the echo of the main sound pulse,

but also for the echoes from the cellars surrounding the various joints of

tubing in the well. 31 By counting the number of tubing collar echoes

before the main echo from the oil surface, and knowing the distance

between each pair of collars, Walker could determine very accurately the

29, See the Ninth Circuit's opinion in Halliburton Oil Well Cementing Co. v. Walker, 146 F.2d at 81"/, 818-20 (9th Cir, 1944).

30. Prior workers in this fiel~J~had used the total return time of the echoes as a measure of the distance down to the pool of oil, erroneously assuming that the velocity of sound was constant throughout the well. This listening-for-the-echo in the prior technology was done electrically, with a microphone, amplifier, and recorder. The error in assuming the velocity of sound often led to misplacement of the production robing and pump, One pair of workers, Lehr and Wyatt, had noticed the error in figuring the velocity of sound in wells, and had devised a corrective formula by which they tried to calculate the veloci:y in each particular well. Interestingly, this was admitted by Walker (see Brief for Cranford P. Walker, et al., Respondents at 8, Hallibunon Oil Well Cementing Co. v. Walker, 326 U.S. 696 (1946) (No. 290)). but not urged by Halliburton, who characterized Lehr and Wyatt as a straight time measurement system, and sought to win the case by urging that the metes and bounds of Walker's claim were indefinite due to the use of functional language at the point of novelty. See Brief for Hallibunon at 4-5, 9-10, id. Walker contended that even the correctional formula of Lehr and Wyatt did not lead to accurate results.

31. Brief for Cranford P. Walker, et al., Respondents at 8-9, M.

162 Harvard Journal of Law & Technology IVol. 8

distance f rom the top o f the well to the surface of the oil pool. 32 If

necessary, the pump ' s location then could be adjusted by raising or

lowering the string o f tubing to which the pump was attached.

Walker implemented his scheme in two versions. In one, he inserted

into the top o f the well a small tube, whose length he made adjustable so

as to be part icularly recept ive to the echoes coming from the many

successively deeper tubing collars in the well 33 and relatively unreceptive

to other frequencies. He called this a resonator tube 3a and it was

described in his '519 "acoust ic" disclosure patent. His c la im to a depth-

measuring apparatus, in addition to recit ing a sound source and a receiver

for picking up echoes, inclt, ded the fol lowing element: "means . . . for

tuning said receiving means to the frequency of echoes from the tubing

lcoupling] collars . . . to clearly distinguish the echoes from said

couplings f rom each other. "3s

Walker realized that an analogous kind of signal-accentuation could be

done electr ical ly. 36 He filed a separate patent application for the electrical

version. 37 His electrical echo-processor c la im was rejected by the Patent

Office for being an obvious equivalent of his acoustic version. The

application with acoustic disclosure meanwhile issued and became the

patent in suit in Halliburton.

B. Walker's "Means For" Claim Is Held Indefinite

Hall ibur ton 's accused infringing system was an electrical version. 3s

Hall iburton contended it was free to use an electrical filter in its echo-

processor structure. 39 Walker o f course disagreed. ~°

32. Walker's scheme required some means to delineate sharply the ~ltber weak echoes from the tubing collars, so as not to miss any of them. It was known at the time that such echoes existed, and that the frequency of each such echo was a function of the annular height of the open space outside an individual joint of tubing in the well (typically a tubing joint was twenty feet long at that time), ld.

33. The second harmonic would arrive at the surface simultaneously with the fundamental and other harmonics from the respective tubing collar.

34. Halliburton Oil Well Cementing Co. v. Walker, 329 U.S. 1, 7 (1946). 35. ld. at 9, n.7. 36. This he did by inserting a microphone into the well, electrically filtering the

microphone's output signal to accentuate a harmonic of the tubing-section-echo signals, and using the filtered electrical signal to drive the same kind of stylus he was using in the acoustic version of his invention.

37. Brief for Halliburton at 38-39, Halliburton Oil Well Cementing Co. v. Walker, 329 U.S. 696 (1946) (No. 290).

38. Halliburton Oil Well Cementing Co. v. Walker, 329 U.S. 1. 121 (1946). 39. Brief for Halliburton Oil Well Cementing Co.. Petitioner, On Reargument Before a


The district court ~t and the Ninth Circuit 4" found Walker's '519

acoustic-disclosure patent valid and infringed. The appellate court relied

on the equivalency between Walker's use of acoustical resonance and

Halliburton's use of electrical resonance for the same purpose. "~3 There

was never any contention by Walker that his "means" expression covered

all means for accomplishing the stated function, but rather only those

structures equivalent to what his specification showed. The Ninth

Circuit's decision was grounded on the view that Halliburton was using

not just any equipment for determining depth by listening for and counting

tubing collar echoes, but equipment which in the context of Walker's

discovery about tubing-collar echoes was the full equivalent of the tuning

pipe disclosed in Walker's specification. 4~

On certiorari, the Supreme Court at first affirmed per curiam by an

equally divided court. ~s A petition for rehearing was granted, 46 resulting

in a reversal of the Ninth Circuit's decision and a holding that Walker's

patent claim was invalid. 47 The ground was solely a supposed indefinite-

ness of the phrase: "means associated with said pressure responsive

device for tuning said receiving means to the frequency of echoes from

Full Bench at 47-85, Halliburton Oil Well Cementing Co. v. Walker, 329 U.S. 1 (1946) (No. 24). Halliburton contended primarily: (a) such a device was no: equivalent to Walker's disclosed acoustical resonator tube; (b) HaUiburton's use of a known electrical filter to accentuate tubing-coupling echoes was merely an obvious design expedient drawn from the prior art and not from Walker's work; and (c) Walker. despite efforts to do so in the Patent Office. had been unable to obtain allowance of his electrical-filter claim, which would have squarely covered the accused Halliburton arrangement.

40. Walker contended that: (a) at trial he had proved equivalence between his disclosed acoustical resonator and Halliburton's electrical filter, both used to accentuate tubing-collar echoes in order to obtain a more accurate distance measurement to the surface of the downbole oil, as conceived by him; (b) the Patent Office had regarded acoustical and electrical filtering to be so obviously equivalent in the context of Walker's invention that it had refused Walker's electrical application in light of his acoustic application; and (c) since the language of the claim in suit was clear, was no broader than Walker's contribution to the art, and squarely embraced Halliburton's electrical structure, it would be unjust to restrict the claim's coverage to merely the acoustical embodiment disclosed in the Walker specification. Brief on Rehearing for Cranford P. Walker, et al., Respondents at 28-36, Halliburton Oil Well Cementing Co. v. Walker, 329 U.S. 1 (1946) (No. 24).

41. Walker v. Halliburton Oil Well Cementing Co., 59 U.S,P.Q. (BNA) 179 (S.D. Cal. 1943).

42, Halliburton Oil Well Cementing Co. v. Walker, 146 F.2d 817, 818-21 (9th Cir. 1944), reh'g denied, 149 F.2d 896 (9th Cir. 1945).

43. Id. at 819-21. 44. Id. 45. Halliburton Oil Well Cementing Co. v. Walker, 326 U.S. 696 (1946). 46. Halliburton Oil Well Cementing Co. v. Walker, 327 U.S. 812 (1946). 47. Halliburton Oil Well Cementing Co. v. Walker, 329 U.S. I (1946).


tile tubing collars . . . . ,,4~

C. Analysis of the Case

Walker ' s brief before the Supreme Court on rehearing makes cleat that

he sought to cover some range of equivalents of his tuning pipe. His

lawyers argued that it was the "intention of both Walker and the Patent

Office that ".his patent should cover not only the tuning pipe 18 [of the

specification] but also other equivalent mechanisms which cotiid be used

for the purpose of carrying the patented invention into practice. ,49

However, Walker was not endeavoring to have his claim cover all

possible means for tuning the receiver to listen for tubing-collar echoes.

He appears to have been urging only that his "means for" expression

should be read to cover his specific "tuning pipe 18" plus equivalents of

that specific structure. This would presumably be a significantly smaller

class of things than "all means" for tuning, due to the requirement

inherent in the term "equivalent" that any covered structure must work

in substantially the same way as an acoustic tuning p i p e : ° Thus Walker

was contending for a ruling virtually identical to the present wording of

112(6). The Supreme Court nevertheless accused Walker of attempting to

cover all possible means for carrying out the recited function:

Un&.r these circumstances the broadness, ambiguity, and

overhanging, tlareat of the functional claim of Walker

becomes apparent. What he C!a~,med in the court below and

what he claims here is that his patent bars anyone from

using in an oil well any device heretofore or hereafter

invented which combined with the [previously known] Lehr

and Wyatt mac.SAne performs the function of clearly and

distinctly catching and recording echoes from tubing joints

with regular i ty) ~

48. /d. at 8-9, 12. 49. Brief on Rehearing fer Cranford P. Walker, et al., Respondents at 29, Halliburton Oil

Well Cementing Co. v. Walker, 329 U.S. 1 (1946) (No. 24). 50. Equivalents in patent law are things tl:at work in substantially the same way to achieve

substantially the zame result. See Graver Tank & Mfg. Cu. ",. Linde Air Prods. Co., 339 U.S. 605,608 (1950).

51. Halliburton, 329 U.S. I, 12 (1946).


The Court was concerned that such a reach--which it does not aPPear

Walker was seeking at all--would stifle future creativity:

"j

in this age of technological development there may be many

other devices beyond our present information or indeed our

imagination which will perform that function and yet fit

these claims. And unless frightened from the course of

experimentation by broad functional claims like these,

inventive genius may evolve many more devices to accom-

plish the same purpose. 5-"

The Court 's concern was with the major philosophical issue of the

patent laws: since a specification is often written narrowly and the claims

broadly, how much legal scope should be accorded to such claims when

the courts are called upon to interpret them? This is a question that has

yet to be addressed adequately by Congress or answered by the courts.

Unfortunately, the problem was obscured in Halliburton by the Court 's

grounding of its decision on a supposed indefiniteness of functional

language. There was no sound reason to think that Walker's phrasing

was indefinite.

The real prol:~lem suggested by the Court 's language was the difficulty

of determining w h e n patent exclusivity goes too far in breadth and

therefore provides a disincentive for later research. Should the claim (and

hence the patent grant) be limited only by the prior art, i.e., valid so long

as it is not so broad as to cover appar~.;.us appearing in, 53 or obvious

from, 54 prior work? Or should the claim be read to have a breadth in

some way proportional to the particulars of the embodiments shown in the

specification and drawings? Rather than tackle this formiSable question,

the Court grounded its decision on a supposed inaccuracy in the breadth

of Walker 's claim: "Had Walker accurately described the machine he

claims to have invented, he would have had no such broad rights to bar

the use of all devices now or hereafter known which could accent

52. /d. 53. A patent claim is invalid for anticipation under.35 U.S.C. § 102 if a single

embodiment within its scope appears in prior literature or products. Vas-Cath, Inc. v. Mahurkar, 935 F.2d 1555. 1562 (Fed. Cir. 1991); In re Lukach, 442 F.2d 967, 970 (C.C.P.A. 1971).

54. Obviousness of any one embodiment of the claim in light of the prior art at the time the claimed invention was made will invalidate the claim under 35 U.S.C. § 103. In re

Klein, 987 F.2d 1569, 1570 (Fed. Cir. 1993). ~

166 Harvard Journal o f Law & Technology [Vol. 8

waves."55

The Court ' s ultimate conclusion that "the claims before us . . . fail

adequately to describe the alleged invention ''56 was only a temporary

shelter by which the Court could avoid the far more vexing and pervasive

question of modern patent jurisprudence, viz., how broad should the

scope of a patent be?

D. Congress Acts

The Congressional attempt in 112(6) to address in some way the

unfairness of ltalliburton was a step in the right direction. Means-plus-

function language in defining combination inventions no longer was to be

per se indefinite, hence the precise basis for the Halliburton decision was

overruled, But the quid pro quo for the new statutory permission was to

be substantial: such expressions in patent claims were not to be accorded

their full colloquial scope, but rather would be limited to structures shown

in the disclosure portion of the patent and the equivalents of such

structures.

The issue of how the constraint of 1 12(6) plays out must be considered

from several different perspectives:

(a) In interpreting a claim to see if it impermissibly

embraces versions of the invention which are not

"statutory" subject matter, i .e., versions which are

not eligible for patent protection, 57 does the restrictive

second part of 112(6) operate to hold the claim to

permissible subject matter and thereby save its

patentability?

(b) In interpreting a claim to see if it impermissibly

includes a version of the invention which is not

new, 58 or which was obvious at the time of the

invention, 59 does the restrictive part of 1 12(6) help to

55. Halliburton, 329 U.S. I, 13 (1946). 56. Id. at 14. 5'7. For example, mental processes are non-statutory See, e.g., In re Prater, 415 F.2d

1393 (C.C.P.A. 1969). 58. See 35 U.S.C. §§ 101, 102 (1988) (limiting patentability to new machines, processes,

etc.). 59. See 35 U.S.C. § 103 (1988) (prohibiting patentability of inventions which would have

No. 1] Crisis in Patent Coverage

save the claim?

167

(c) In interpreting a claim to see whether it has been

infringed, does the constraint of 112(6) operate to

render an accused structure noninfringing even

though it contains some sort of "means for" doing the

function recited in the claim?

Succeeding sections hereof will separately address each of these facets of

112(6).

III. APPLICABILITY OF 112(6) TO STATUTORY

SUBJECT MATTER ISSUES

:, A. Iwahashi and Its Aftermath

For purposes of defining statutory subject matter the question of

whether a means-plus-function expression is subject to the narrowing

influence of 112(6) was seldom addressed prior to 1990. One case that

addressed this, though only in dictum, was In re Bernhart. 6° This was

one of the early cases treati~gihe question of statutory subject matter with

reference to functions implementable on general purpose digital comput-

ers. In concluding that the mechanical drawing apparatus 6~ was statutorily

eligible for patent protection, the court in Bernhart observed that 112(6)

limited the coverage of the claim's "means" expressions to "only such

mechanical drafting means and their equivalents. "62 The claim did not

embrace mental implementations, and was therefore statutory.

The issue of applying 112(6) to statutory subject matter decisions was

not thoroughly addressed by the courts until the 1990 decision of the

Federal Circuit in In re lwahashi . 6~ The case involved data-processing

equipment for voice-recognition use. ~The high ~peed apparatus was

designed to carry out a mathematical function known as auto-correlation.

On appeal from a PTO rejection was an apparatus claim for a combina-

been obvious to a person of ordinary skill when made). 60. 417 F.2d 1395 (C.C.P.A. 1969), ~ , 61. The claimed apparatus plotted mechanical drawings automatically based.~;: input

dimensional data about the object being drawn, ld. at 1396-97. 62. Id. at 1399. 63. 888 F.2d 1370 (Fed. Cir. 1990).

168 H a r v a r d Journa l o f L a w & Techno logy [Vol. 8

t ion o f e lements , all bu t one o f wh ich (the read-only m e m o r y ) were

expressed in means -p lus - func t ion language. The PTO Board o f Appeals

had re jec ted the c l a im unde r Got t scha lk v. Benson 64 and its p rogeny , 65 as

essent ia l ly p r e e m p t i n g a mathemat ica l a lgor i thm and hence be ing d rawn

to nons ta tu tory subjec t mat te r .

The Federa l Ci rcu i t reversed , r eason ing that the c la im def ined

re la t ionships a m o n g s t ructura l componen t s ra ther than a mathemat ica l

a lgor i thm. ~ Af te r s ta t ing its holding, the cour t took issue wi th a

s ta tement in the PTO Sol ic i to r ' s b r i e f to the effect that the c l a im on

appeal " e n c o m p a s s e s any and eve ry means for p e r f o r m i n g the funct ions

reci ted therein."67 The court , not ing that the c la im was for a combina t ion

of parts all but one o f which were in means-p lus - func t ion t e rmino logy ,

stated: " T h e c l a im is there fore subject to the l imita t ion stated in 35

U . S . C . 112 para . 6 that each means -p lus - func t ion def ini t ion ' shal l be

cons t rued to cove r ,the co r r e spond ing s t ructure , mater ia l , or acts descr ibed

in the speci f ica t ion and equivalents thereof . ' "6s )

Immedia te ly fo l lowing this passage came a footnote, 69 where in the

64. 40-9 U.S. 63 (1972). 65. See infra note 79. 66. See also In re Walter, 618 F.2d 758 (C.C.P.A. 1980) (holding that a claim involving

a mathematical algorithm could be statutory if the algorithm is implemented in a specific manner to define structural relationships among the physical elements of the claim). Id. at 767.

67. lwahashi, 888 F.2d at 1375. 68. ld. 69. The famous "footnote I ~ in lwahashi read:

The accuracy of this statement may be questioned in view of a sentence in the opinion in In re Sweet, 393 F.2d 837, 841-42, 157 USPQ 495, 499 (CCPA 1968), which reads: "[A] recitation of 'means' [sic] for performing a function is interpreted broadly to cover all means capable of performing the stated function and is not limited to the particular structure which the application may disclose." (Emphasis added.) This statement, considered in a vacuum, is partly true and partly untrue. It must be read, however, in light of the opinion as a whole. It should not be removed from its context. The untrue part is the initial statement that the means clause is interpreted to cover all means to perform the function. It should have said it is interpreted to cover the means disclosed and all equivalents thereof which perform the function. The immediately preceding two paragraphs of the opinion show that the cour~ actually was reading the "means ~ clause "in the light of 35 U.S.C. 112 [last paragraph]," just asiwe are doing here. The statute is set forth in note 5. The truth of the emphasized portion of the above sentence is beyond question because the "means" clause includes equivalents of the disclosed structure. Section 112 ¶6 cannot be ignored when a claim is before the F r o any more than when it is before the courts in an issued patent.

No. lI, Crisis in Patent Coverage 169

court undertook to explain away a passage in In re Sweet 7° which seemed

to say that means expressions should be interpreted broadly to cover a!!

means for achieving the recited function.

This "footnote 1" of lwahashi has come to be rather well known.

Less than three months after the decision in lwahashi, the PTO took pains

to announce that the court 's opinion therein did not alter the PTCYs view

of statutory subject matter under Section I01: in judging whether a claim

impermissibly embraced nonstatutory subject matter, a "means for"

expression would be deemed by the PTO examiner to cover all means for

carrying out the recited function. 7I Acting Assistant Commissioner Denny

issued a notice stating that the court 's view of such expressions, as recited

in footnote 1 of lwahcL~;'.i, was "directly contrary to precedent. ''72 For

PTO examination purposes such expressions would not be impacted by

112(6), Would be deemed to cover any and all means for carrying out the

recited function, and hence would be more likely to be rejected as

embracing nonstatutory subject matter.

The PTO was correct that there did not appear to be any logical basis

to differentiate claim scope for purposes of prior-art comparison 73 from

claim scope for purposes of judging whether the claim embraced any

nonstatutory subject matter. The problem was whether either issue should

be decided on bases other than as expressed in 112(6) and, more

generally, whether pre-issuance PTO proceedings to obtain a patent

should be conducted on criteria which are different from the criteria used

by the courts in reaching post-issuance judgments on that same patent, the

statutory provisions governing both types of proceedings being the same.

B. The Seeming End of the Road: Arrhythmia

The Federal Circuit has had one more opportunity to pass on this:

contentious issue, although it was in a case to which the PTO was not a

party. In Arrhythmia Research Technology, Inc. v. Corazonix Corp., 74

the "ourt dealt with ;~n ;.ppeal from a district court infringement litigation

hi. at 1375 n.1 (punctuation and citations as In original). 70. 393 F.2d 837, 841-42 (C.C.P.A. 1968). 71. The claim would thus be more likely to be rejected as embracing forbidden subject

matter. 72. James E. Denny, Notice Interpreting hz re lwahashi, 39 Pat. Trademark & Copyright :~-~

J. (BNA) 399 (199¢0, reprinted in l 112 OFF. GAZ. PAT. OFFICE 16 (Mar. 13, 1990). 73. See infra part IIl. 74. 958 F.2d 1053 (Fed. Cir. 1992).

170 Harvard Jottrnal o;~ Lmv & Technology [Vol. 8

J : .

decis ion gr~,lamg s u m m a r y j u d g m e n t o f inval idi ty o f a patent on the

g round o f nons ta tu tory subject mat ter . 75 The invent ion involved a method

and compu te r i zed appara tus for ana lyz ing a hear t -a t tack pa t i en t ' s EKG,

to d e t e r m i n e whe the r or not the pat ient was prone to tachycard ia and

ven t r i cu la r f ibr i l la t ion in the hours immedia te ly af ter the at tack.

Some of the c la ims o f the patent in suit were to the method of

ana lyz ing the EKG; o thers were to the appara tus for doing so. The

appara tus c la ims were in means -p lus - func t ion language, 76 and the distr ict

court had s u m m a r i l y held them invalid for fai lure to c la im statutory

subject mat ter . 77

The appel la te court , in a two-sen tence pa ragraph ci t ing lwahashi, held

that in v iew of the language o f ! 12(6) " the s ta tutory nature vel non of

[the pa ten tee ' s ] c la ims is de t e rmined with re fe rence to the descr ip t ion in

the ... specif icat ion."78 The cour t p roceeded to rule that the specif icat ion

75. Id. at 1054. 76. The cot,-rt treated claim 7 as typical, ld. at 1055. The claim read as follows:

7. Apparatus for analyzing electrocardiograph signals to determine the level of high frequency energy in the late QRS signal [a part of the heart's electrical cycle, in which high-frequency energy at a low voltage level is an indicium of susceptibility to tachycardia if] comprising:

means for converting X, Y, and Z lead electrocardiographic inpiat signals to digital valued time segments;

means for examining said X, Y, and Z digital valued time segments and selecting therefrom the QRS waveform portions thereof;

means for signal averaging a mul¢iplicity of said selected QRS waveforms for each of said X, Y, and Z inputs and providing composite, digital X, Y, and Z QRS waveforms;

high pass filter means;

means for applying to said filter means, in reverse time order, the anterior portion of each said digital X, Y, and Z waveform; and

means for comparing the output of said filter means with a predetermined levei to obtain an indication of the presence of a high frequency, low level, energy component in the filter output of said anterior portions.

Id. The central novelty was said to be in the reverse-order filtering of the late QRS sample signals, ld. The PTO had granted this claim without questioning whether it was statutory. Id.

77. Although the language of 112(6) expressly applies to a "means or step," the court did not mention 112(6) in its treatment of the method claims. See id. at 1058-61.

78. Id. at 1060.


did not address any abstract mathematical solutions or logical algorithms 79

but rather a heart-attack patient's EKG, a non-abstract technological area

clearly within the reach of the patent laws. s°

The PTO saw in Arrhythmia no reason to depart from the contrary

views of 112(6) expressed in the "Denny Notice ''st and continued to

ignore 112(6) in interpreting means-for expressions in claims before the

PTO. The struggle between the court and the PTO on this issue was to

escalate ~lramatically.

C. The Commissioner Enters the Fray: Alappat

Up :mtil this time, the PTO's views had not been expressed by the

Commissioner personally. The notice by Assistant Commissioner Denny

had been in rather routine form. Now, however, the stage was set for the

Commissioner s,- to assert a position contrary to the court's, and to do so

by intruding personally upon the work of the Board of Patent Appeals and

Interferences.S3

The case in which this happened was Ex parte Alappat. s4 A three-

L

i ,

79~. Coverage of a logical algorithm per se, unrelated to a particular end use, was found nonstamtory in Gottschalk v. Benson, 409 U.S. 63, 72 (1972). The ruling has been the source of extensive commentary and refinement in later cases. See, e.g., Parker v. Flook, 437 U.S. 584, 591 (1978) (holding that preemption of algorithm is one basis for finding a claim nonstatutory, and suggesting that a mathematical algorithm in a claim should be disregarded in reaching the conclusion on its statutory material); Diamond v. Diehr, 450 U.S. 175, 185 (1981) (finding rubber-curing claim statutory even though it involved algorithm for solving Arrhenius' equation); In re Freeman, 573 F.2d 1237 (C.C.P.A. 1978) (establishing a two-part test to determine whether the claim is statutory: (I) seeing if claim involves algorithm; (2) determining whether claim as a whole is no more than the algorithm itself); hz re Walter, 618 F.2d 758 (C.C.P.A. 1980) (same); Donald S. Chisum, The Patentability of Algorithms, 47 U. PITT. L. REV. 959 (1986) (arguing that Diehr in part superseded Benson and Flook.)

80. Arrhythmia, 958 F.2d at 1060-61. The court noted that the final output of the apparatus was a measure, in microvolts, of voltage at the heart's ventricle. Id. at 1060.

81. See supra note 72. 82. The Commissioner also serves as Assistant Secretary of Commerce. See 35 U.S.C.

§ 3(d) (1988). 83. Pursuant to 35 ~i.S.C. § 7(a) (1988), this board consists of the Commissioner, the

Deputy Commissioner and the two Assistant Commissioners, all of whom are Presidential appointees with advice and consent of the Senate, 35 U.S.C. § 3(a) (1988), plus the examiners-in-chief, who are appointed by the Secretary of Commerce on the nomination of the Commissioner. ld.

84. 23 U.S.P.Q.2d (BNA) 1340 (Bd. Pat. App. & Interferenc_-':--i992). An appeal from the decision was taken en banc by the Federal Circuit. See In re Alappat, 980 F.2d 1439 (Fed. Cir. 1992). The court directed the parties to brief various issues regarding the Commissioner's authority: Could the Commisioner augment the panel's membership and subsequently order a rehearing after the panel had rendered a decision contrary to the PTO's

172 Harvard Journal o f Law & Technology lVol. 8

member panel o f the Board decided the case a5 in the applicant 's favor by

applying 1 12(6) in accord with the Federal Circui t ' s interpretation thereof

in lwahashi. The Commiss ioner determined that a rehearing should be

granted and that he and four other senior PTO officials should be added

to the panel, 86 The newcomers became the majori ty, changing the

decision in the case by adhering to the PTO view that 112(6) does not

apply in PTO proceedings. The three original panel members dissented,

urging that the Federal Ci rcui t ' s v iew, not the P T O ' s view, on means-

plus-function claims was the law of the land.

Alappat concerned a patent application involving a "ras ter izer" for

sampling an incoming electrical signal frequency and recording internally

t!~ m~gnitudes sampled. These internal data can then be manipulated to

produce a visually displayed wavefo rm that is a modified foi'm of the

incoming signal, correc ted for aberrations that may have corrupted the

incoining signal before it was received, a7 The claim in issue called for a

combinat ion which included a "means for" looking up point locations in

a table created from incoming, non-standardized data, then adjusting the

position data to fit the nearest available standardized locations on the

screen, and finally displaying the set o f new points as a smooth line. ss

The apparatus disclosed in the applicat ion 's specification involved

ari thmetic logic units to determine the standardized point positions, s9

The Board majori ty took the v iew that all the functions listed in the

"means for" recitations o f the claim could bc done in a p rogrammed

:J official view on 112(6)? The decision of the Board was rever~ed. See b~ re Alappat, 33 F.3d 1526 (Fed. Cir. 1994).

85. Decisions by three-member panels are authorized by 35 U.S.C. § 7(b) (1988) and are the normal way that most appeals are disposed of by the PTO. The Commissioner or his delegate does have the power, however, to order a larger group of the Board to hear the appeal. Section 7(b) allows the Commissioner to appoint additional members of the Board to hear a given appeal, prov!ded that "at least three members of the Board" are so designated. In some instance;,; to maintain timeliness of the PTO work product, the Commissioner can designate z', lower-ranking PTO official to sit on a Board panel. See 35 U.S.C. § 7(c) (1988).

86. The appointees were: Commissioner Manbeck, Deputy Commissioner Comer, Assistant Commissioner Samuels, Board Chairman Scrota, and Board Vice Chairman Calvert. E.rparte Alappat, 23 U.S.P.Q.2d (BNA) at 1340.

87. The court characterized this as "an electrical device that converts waveform magnitude data in~.o an array of intensity data for use in creating a smooth waveform display." Id. at 1350.

88. See M. at 1341 (claim 15). 89. The inventor did not provide further detail about how the standardized point positions

are determined. The majority opinion mentioned a barrel shifter for doing the normalizing, ROMs for outputting data, and a table-lookup means, in addition to the arithmetic logic units. Id. at 1344.


digital computer, and that 112(6) did not operate to limit the meaning of

the claim terminology in any wayfl ° Thus read without regard to 112(6),

the majority went on to hold that the claim effectively preempted a logical

algorithm and hence was nonstatutory under the Supreme Court case law

described above. 9t

The Board majority took express note 9-" of the Federal Circuit 's recent

decision in Arrhythmiaf l 3 but sought to distinguish it on the grounds that:

(1) the claims in Arrhythmia were more specifically drawn; 94 and (2)

Arrhythmia was an appeal from an infringement action involving a patent,

and the rules of claim construction in infringement ~,ctions are different

than the rules for claim interpretation during ex parte prosecutionfl 5

Before the Commissioner ordered reconsideration by the expanded

panel, the Board members who decided the appeal had, of course, come

out the other way. Now, they expressed considerable indignation over the

new majori ty 's opinionfl 6 They asserted that In re lwahashi 97 stood for

"the proposition that claims drafted in 'means for' format are to be

construed in light of the disclosed means for performing the functions and

the equivalents thereof in accordance with the sixth paragraph of 35

U.S.C. 112 in determining whether they are statutory or nonstatutory

under 35 U.S.C. 101."98

The stage was now set for the Federal Circuit to state definitively

whether means-plus-function claims were to be interpreted more broadly

in a PTO proceeding than in district court actions.

D. The Federal Circuit Responds:

One Harmonious Chord in a Dissonant Sea

On appeal, the Federal Circuit took the case en banc, obviously

distressed by the Commissioner ' s apparent "stacking" of the panel with

90. /d. at 1345. The board majority specifically stated that the "'means' of claim 15 read on any and every means for performing the functions." Id.

91. Id. at 1347. 92. ld. 93. Arrhythmia Research Technology, Inc. v. Corazonix Corp., 958 F.2d 1053 (Fed. Cir.

1992). See supra part III.B. 94. Er parte Alappat, 23 U.S.P.Q.2d (BNA) at 1347. 95. Id. 96. See id. 97. 888 F.2d 1370 (Fed. Cir. 1989). See supra part III.A. 98. E~ parte Alappat, 23 U.S.P.Q.2d (BNA) at 1349.


members who shared his views on 112(6). ~ However, in deciding the

case the Federal Circuit judges themselves were far from accord on most

of the issues raised by the appeal, issuing seven s!iparate opinions.l°°

The single point of accord among all the jtidges who reached the

merits in Alappat was on the subject of 112(6): all found that the

expanded panel had erred in not applying the limiting effect of 112(6) to

the claims in determining whether they were statutory.m~ In other words,

the PTO could not ignore the statute any more than district courts could.

E. Prehtde to a Peace Accord?

It now appears that with a new Commissioner in office m-' the PTO may

99. See In re Alappat, 980 F.2d 1439 (Fed. Cir. 1992). 100. hz re Alappat, 33 F.3d 1526 (Fed. Cir. 1994) (en bane). On the question of whether

the enlarged rehearing panel was so infirmly constituted that its ruling was not an appealable "decision" at all, Judge Rich, joined by Judges Lcurie, Newman, and Rader, said the panel was properly constituted. Chief Judge Archer and Judges Nies and Plager concurred in the conclusion, ld. at 1530, 1545, 1577. Judges Mayer and Michel dissented, id. at 1571, as did Judges Schall and Clevenger, id.

On the merits of the appeal, Judge Rich ruled that Alappat's claim, properly interpreted per 112(6~,~ was statutory subject matter, ld. at 1530. Judge Newman also issued a concurring opinion. Judge Newman explained that she thought the patent system should I~ open to new technologies and should not be bogged down in squabbles about principles t)f mathematics being outside its reach. Id. at 1568. Judge Rader did the same. Judge Rader reasoned that the Supreme Court's decision in Diamond v. Diehr, 450 U.S. 175 (1981), severely constrained earlier Supreme Court decisions against algorithms being patentable . - : subject matter. An algorithm is not patentable subject matter only where the sole discovery:~ ~ • defined by the claim was a principle of nature itself, ld. at 1581.

Judge Plager also issued a separate opinion, which did not address the patentable subject matter issue, other than to say that the panel erred in its failure to follow 112(6) and that he applauded Judge Rich's "chasing out of some of the less useful accretions regarding patentability under § 101." ld.

Two of the remaining ONe judges, Archer and Nies, dissented on the merits. These judges took the view that tl,,e majority were disregarding the coti:,tl tints on patentable subject matter imposed by prior cases. Id. at 1543. l'he other three judges (Mayer, Clevenger, and Schall) said that in view of the panel',, ..',.aving been improperly constituted there was no decision to review, and accordingly they had no opinion on the merits. See id. at 1571 (Mayer, J., dissenting, joined by Michel, J., on this point); id. at 1583.

101. ld. at 1540 (Rich, J.) (arguing it was error for panel to refuse to apply 112(6) in rendering its § 101 patentable subject matter determination); id. at 1561 (Archer, C.J., concuning and dissenting in part) (agreeing that panel erred in failing to apply narrowir.g constraint of 112(6) ~n patentable subject matter determination but finding the issue a "red herring"); id. at 1568.(Newman, J.+ concurring) (joining Judge Rich on the merits and expressing additional views); id. at 1577 (Plager, J., concurring) ("there is no doubt that the Board erred as a matter of law in refusing to apply § 112 [pam.l 6 as we have instructed"); id. at 1581 (Rader, J., concurring) ("the Board reacheta its conclusion by impermissibly expanding the scope of the claimed subject matter, thereby runniag afoul of 35 U.S.C. § 112, [pam.l 6").

102. Commissioner Bruce Lehman took office August 6, 1993. Legislation: Senate


have ameliorated its hostility to applying 112(6) in PTO proceedings. In

the new Examiner Guidelines published April 28, 1994, the Office

announced that in prior art determinations it would now apply 112(6), in

light of numerous Federal Circuit holdings to the effect that it must do

so. t°3 The literal wording of the new guidelines would also apply to

statutory subject matter determinations, since the guidelines are phrased

in terms of how examiners should interpret functional language in chlims. ~04

Nevertheless, it is not entirely clear whether the PTO has capitulated

to the 112(6) aspect of the Federal Circuit's Alappat decision. As noted

above, all eight judges addressing the issue in Alappat said the narrowing

effect of 112(6) applies to PTO determinations of whether claims are

properly in accord with statutory subject matter requirements.t°s The new

guidelines, issued several months before Alappat was decided, were

promulgated in response to In re Donaldson. ~°6 Donaldson involved only

an obviousness rejection under Section 103. Moreover, the guidelines say

Donaldson affects 112(6) and does not "directly affect the manner in

which any other section of the patent statutes is interpreted or applied."~°7

This passage leaves it unclear whether the new guidelines will impact

PTO determinations of statutory subject matter under Section 101.

The PTO should acquiesce on statutory subject matter determinations

as it has now acquiesced on prior art determinations. That result would

Confirms Bruce Lehman Jbr Commissioner of Pate, its and Trademarks, 46 Pat. Trademark & Copyright J. (BNA) 311 (1993).

103. For the text of the new guidelines, see Charles E. Van Horn, PTO Notice on Means or Step Plus Function Limitation Under 35 U.S.C. Section 112, 6th Paragraph, 47 Pat. Trademark & Copyright J. (BNA) 571 (1994). The PTO intends to incorporate the guidelines into the Manual for Patent Examining Procedure. hi. See blfra part IV (discussing cases which led to the new guidelines involving 112(6) in determinations of patentability over the prior art).

104. The guidelines recite in part:

The purpose ot this memo is to set forth guidelines for the examination of § 112, 6th paragraph ~means or step plus function" limitations in a claim . . . . [Elffective immediately, examiners shall interpret a § 112, 6th paragraph "means or step plus function" limitation in a claim as limited to the corresponding strucn~reo materials or acts described in the specification and equivalents thereof in accordance with the following guidelines.

~d. 105. See supra note 101 and accompanying text. 106. 16 F.3d 1189 (Fed. Cir. 1994). 107. Van Horn, supra note 103 at 571. Footnote 10 to the new guidelines mentions the

statutory subject matter context of 112(6) but does not specify whether examiners are to treat means expressions in the 112(6) manner when judging patentable subject matter.

176 Harva,'d Journal o f L a w & Technology [Vol. 8

be far better than perpetuating the strain between the court and the PTO. JoB

IV. APPLICABILITY OF THE STATUTE TO

NOVELTY AND OBVIOUSNESS

DETERMINATIONS

Whether 112(6) applies in determining patentability over the prior art

under Sections 102 and 103 has also been the subject of significant debate

and disagreement between the PTO and the Federal Circuit. This conflict

has perhaps been even more rancorous than the disagreement over the

applicability of 112(6) to statutory subject matter determinations.

The issue, at least for purposes of judging patentability of a claim over

the prior art, is whether a means-plus-function expression in a patent

claim reads on all means for achieving the function, or whether 112(6) ~°9

causes such a recitation to be construed as embracing only the structures

shown in the specification plus their equivalents. In In re Bond tt° the

Federal Circuit he ld that 112(6) applied to such a patentability determina-

tion. The invention involved a delay circuit used in connection with

telephone-answering machines. The claim under consideration in the PTO

recited "delay means ." The specification showed digital delay circuitry

and a prior-art reference showed analog circuitry, both for performing

what the PTO viewed as the recited function in the claim. Itt The court

held that 112(6) applies to novelty determinations: "While a 'means-plus-

funct ion ' may appear to include all means capable of achieving the

desired function, the statute [112(6)] requires that it be 'construed to

cover the corresponding structure, material, or acts described in the

108. Unfortunately there are signs that Commissioner Lehman will follow the practice of some of his predecessors in declining to follow certain Federal Circuit decisions which are handed down by a three-judge panel and deemed by the PTO to conflict with prior authority. See PTO Will Not Follow CAFC Decision on Obviousness of Chemical Compounds. 47 Pat. Trademark & Copyright J. (BNA) 500 (1994) (indicating that Commissioner Lehman had issued an internal PTO memorandum instructing examiners to disregard In re Baird, 16 F.3d 380 (Fed. Cir. 1994) (holding that a chemical compound was not prima facie obvious merely because it was embraced within a generic claim of a prior U.S. patent), as being out of accord with prior cases).

109. For ehe text of 112(6), see supra text accompanying note 13, 110. 91,:'~ : 2 d 831 (Fed. Cir. 1990). 111. Id :.~ 53;2-34. The court was unsure about whether the prior art reference had any

structure tt, accomplish the recited delay function. The court remanded . .r a determination by the PTO on that point. Id. at 835.


specification and equivalents thereof.'"1 u

The court cited In re lwahashi m for the proposition that 112(6) applies

equally to court and PTO determinations, TM and Johnston v. IVAC

Corp. ,5 for the proposition that "Section 112 para. 6 operates to cut back

on the types of means which could literally satisfy the claim language."~6

The PTO board performed no equivalency analysis. Its anticipation

finding was vacated and the case remanded for an equivalency analysis. 117

Reaction from the PTO came two years later. In Ex parte tsa!;sen, tt~

decided by an expanded Board panel including both Commissioner

Manbec and Chairman Serota, the panel expounded at length on the

PTO's vit. that 112(6)was inapplicable to PTO proceedings; that

since an applk. -,t before the PTO, unlike a litigant in the federal courts,

can readily amend his claims, there is no reason to permit reliance on

112(6); that the history of 112(6) as a response to the validity and

infringement concerns of the Supreme Court in Halliburton signifies that

i t is inapplicable to PTO proceedings; that commentaries at the time of

enactment indicate that 112(6) was not intended to operate on patentability

questions, and especially not before the Patent Office; that authority prior

to Bond was nearly uniformly the other way; and that a host of perceived

practical problems would result from the PTO's following B o n d , If9 The

panel concluded that in PTO proceedings regarding patentability, unlike

court p-oceedings regarding validity, means-plus-function expressions ar,~

deemed to include all possible means for carrying out the recited

function. P.0 Six days later, the Commissioner released a detailed Notice

setting forth the same analyses and positions, x'J The painstaking detail of

the PTO's analysis suggests that the PTO took the issue very seriously.

112. /d. at 833. 113. 888 F.2d 1370. 1375 n.1 (Fed. Cir. 1989). See supra notes 60-72 and accompanying

text, 114. See also Carl H. Moy, The bzterpretatiot, of Means E~pressions During Ftamination,

68 J. PAT. OFF. SO¢'Y 246 (1968). lwahashi was actually a statutory subject matter case under Section 101.

115. 885 F.2d 1574 (Fed. Cir. 198t~) 116. In re Bond, 910 F.2d at 833 (citing Johnston, 885 F.2d at 1580). 117. /d. 118. 23 U.S.P.Q.2d (BNA) I001 (Bd. Pat. App. & Interferences 1991). 119. Id. at 1007-15. 120. ld. at 1015. The panel stated as to the case before it: "Any means capable of

performing the recited function meets the claim due to the scope of pending means claims discussed infra." ld. at 1096.

121. Harry K. Manbeck, Jr., PTO Notice On Application of 35 U.S.C. l l 2 para. 6, 43 Pat. Trademark & Copyright J. (BNA) 161 (Dec. 19, 1991), reprinted in I134 OFF. GAZ. PaT. OFF. 631 (Jan. 7, 1992).

178 H a r v a r d Journa l o f L a w & Technology lVol. 8

The Federal Circuit in February, 1994 responded en banc with In re

Dona ldson Co. 122 With Alappa t still pending before the court en banc to

resolve Section 101 patentable subject matter determinations, Donaldson

squarely addressed the PTO's position of disregarding the limiting effect

of 112(6) whel~ judging patentability over prior art under Sections 102 and

103. The case involved re-examination of an air-cleaning systems patent.

The Examiner had rejecled the claim on the ground of obviousness, and

the Board affirmed. 123

The patent owner, Donaldson, effectively conceded that a prior art

reference disclosed every element of the claim but one. 1'-4 That element

was written as a "means" expression, for moving particulate matter

(typically dust) in a downward direction toward a collecting hopper, in

response to pressure changes in a dirty-air filtering chamber.125 The prior

art taught the use of air-pressure pulses to dislodge dust stuck to the

outsides of air-filter elements. However, Donaldson's patent disclosed in

its specification and drawings a flexible-walled arrangement for the

collecting hopper. The hopper thus acted like a diaphragm, expanding in

response to periodic air pulses, and returning to its original shape after

each air pulse.Z26 This avoided the prior problem of dust and debris

caking on the walls of the hopper and becoming difficult to remove.

The court pointed out that this kind of language is read according to

the restrictive mandate of 112(6):

The plain and unambiguous meaning of paragraph six is that

one construing means-plus-function language in a claim must

look to the specification and interpret that language in light

of the corresponding structure . . . described therein, and

equivalents thereof, to the extent that the specification

provides such disclosure• Paragraph six does not state or

even suggest that the PTO is exempt from this maudate, and

there is no legislative history indicating that it should be. 127

122. 16 F.3d 1189 (Fed. Cir. 1994). 123. Id• at 1191. 124. Id. at 1192. 125. The claim language in issue read: "a lowermost portion in said filtering chamber

• . . for collection of particulate matter, said portion having means, responsive to pressure bzcreases in said chamber . . , for moving particulate matter in a downward direction to a bottommost point in said portion for subsequent transfer to a location exterior to said assembly." ld. at 1191.

126. Id. at 1190-91. 127. Id. at 1193 (footnote omitted). In a lengthy footnote the court reviewed the legislative

No. 1] Cr is is in Patent Cove rage 179

The cour t then express ly over ru led any cont ra ry p receden t j'-8 and

p roceeded di rect ly to take on the P T O ' s longs tanding prac t ice of ignor ing

112(6), saying: " T h e fact that the PTO may have failed to adhere to a

s ta tutory manda te ove r an ex tended per iod of t ime does not just i fy its

con t inu ing to do SO. "129

Turn ing to the meri ts , the cour t v iewed D o n a l d s o n ' s c la im language

as cover ing , per 112(6), the f lexib le-wal led col lect ing c h a m b e r and its

equiva lents . The pr io r ar t did not show such a feature or render it

obvious ; hence , the cour t r eversed the PT O de te rmina t ion . ~3°

Two months later, the PTO express ly acquiesced to the cou r t ' s ho ld ing

and analys is o f 1 12(6) in D o n a l d s o n . In a set o f guidel ines for examin -

ers TM the PTO acknowledged the impor t of D o n a l d s o n for novel ty and

obv iousness de te rmina t ions , i .e . , that the PT O must apply 112(6) in

reach ing such de te rmina t ions i f means -p lus - func t ion c la im language is

involved. 132 It d i rec ted Examine r s to fol low the new guidel ines "e f fec t ive

immedia te ly . " *33

The P T O has a l igned its guidel ines wi th the cou r t ' s hold ing . This

indicates that the bat t le ove r 112(6) seems to be over , at least on this

point .

history of various revisions to Section 112 after its original pass~ige in 1952, and concluded there was no evidence that Congress intended to ratify the PTO's practice of ignoring 112(6) in its various determinations of patentability. Id. at 1193 n.3.

128. Id. at 1193-94. 129. ld. at 1194. The court also took issue with the PTO's contention in this case that the

court's precedents on 112(6) were in conflict with the ancient patent-law doctrine that claims should be interpreted as broadly as their language will reasonably permit. The court asserted that its decisions had interpreted claims this way, because in light of the statutory constraint of 112(6) it could not reasonably be urged that means-plus-function claim language was unlimited by the structures shown in the application's disclosure. Id. at 1194-95.

130. Id. at 1195-97. 131. Van Horn, supra note 103 at 571. 132. Id. at 571. 133. ld. In these guidelines, issued by the very agency (PTO) which so long resisted

involvement with 112(6), there appears the clearest and most succinct articulation of 112(6) and how it differs from the equitable doctrine of equivalents:

An "equivalent" for the purpose of § 112, 6th paragraph, should not be confused with tile doctrine of equivalents . . . . Section 112, 6th paragraph limits the scope of the broad "means or steps" limitations in a claim to a combination, to the structures, materials and acts described in the specification and equivalents thereof. The doctrine of equivalents equitably expands exclusive patent fights beyond the literal scope of a claim.

Id. at 574 (citing Valmont Indus., Inc. v. Reinke Mfg. Co., 983 F.2d 1039 (Fed. Cir. 1993)). The PTO's phrasing of the difference between these two concepts is elegant.


V. IMPACT OF 112(6) IN NARROWING CLAIMS

FOR INFRINGEMENT ANALYSIS

Application of 112(6) to infringement analysis is probably the most

important commercial aspect of that statutory provision. Early Federal

Circuit decisions, beginning with D.M.I., Inc. v. Deere & Co. 13-1 stressed

that a means-plus-function expression was not limited, for infringement

purposes, 1o the exact arrangement shown in the specification. The court

in D.M.I. said: "To interpret 'means-plus-function' limitations as limited

to a particular means set forth in the specification would be to nullify the

provision of Sec. 1 12 requiring that the limitation shall be construed to

cover the structure described in the specification and equivalents

thereof."135

This was enough to dispose of the issue then before the court, where

the district judge had improperly limited the scope of the clause to only

the structure disclosed in the specification. However, the court went on

with a statement which wrongly hinted that the clause would cover all

means for accomplishing the function: "Patentees are required to disclose

in the specification some enabling means for accomplishing the function

set forth in the 'means-plus-function' limitation. At the same time, there

is and can be no requirement that applicants describe or predict every

possible means of accomplishing that function. "~36 The implication is that

the patentee will nonetheless get coverage for every such possible means.

In any event, there is nothing in the court 's language to suggest that a

"means" expression would not cover all means.

This potentially misleading dictum of D.M.I. was repeated by other

early decisions in the Federal Circuit 's development of the law regarding

means-plus-function claims.137 A case was needed where the facts would

allow the court to state squarely that the permissive part of 112(6) does

not mean that a functional recitation in a claim will be allowed to cover

any and all means for carrying out the recited function. Texas Instru-

ments x38 was such a case.

134. 755 F.2d 1570 (Fed. Cir. 1985). 135. ld. at 1574. 136. ld. 137. See, e.g., P.M. Palumbo v. Don-Joy Co., 762 F.2d 969, 974 (Fed. Cir. 1985); King

Instruments Corp. v. Otari Corp., 767 F.2d 853, 862 (Fed. Cir. 1985), cert. denied, 475 U.S. 1016 (1986).

138. Texas Insruments v. United States lnt'l Trade Comm'n, 805 F.2d 1558, 1562-63


A. The First Signal

Texas btstruments is the first Federal Circuit case to apply the second

clause of 112(6) to take away some of what the first clause gives. In that

case, the court found that the claim did not cover any and all means for

carrying out its recited function. The rationale behind the decison will

now be explored.

The case involved an International Trade Commission ("ITC")

proceeding ~39 to prohibit the importation of hand-held calculators, which

Texas Instruments ("TI") said infringed its famous Kilby patent on

calculators. The Kilby patent enjoyed an effective filing date of 1967,

and the Federal Circuit found that it was a "pioneer patent. ''~4°

The patent claim in issue was to a miniature, battery-powered

calculator, and was defined in means-plus-function terms, viz.: (1) input

means including a keyboard having a single set of number keys; (2)

electronic means for performing arithmetic on numbers entered by the

keyboard, including integrated circuit memory means for storing the

numbers, arithmetic means for doing arithmetic on them, and means for

transferring the numbers from the memory through the arithmetic means

and back again; and (3) means for providing a visual display of the

answer. The claim recited that all of these "means" were contained in a

pocket-sized housing, z4~

In style this was undoubtedly the quintessential means-for claim.

However, by the early 1980s, the technology for performing every one

of the "means" functions had changed dramatically. For example, the

preferred embodiment of the "display" shown in the patent was a thermal

printer that used heat to print dots in the form of numerals on a paper

tape. The allegedly infringing calculators used the more modern--and

very different-- l iquid crystal display ("LCD") . ~42

The administrative law judge of the ITC found that each of the recited

functions was carried out in the accused calculators. Nevertheless, the

judge did not find infringement because the functions were carried out in

radically different ways due to the advanced electronics available at the

(Fed. Cir. 1986). 139. Tariff Act of 1930 § 337, 19 U.S.C. § 1337, empowers the ITC to issue exclusion

orders barring the importation of goods which infringe U.S. patents. Appeal of the ITC's decision in such a case is made to the Federal Circuit. Id.

140. Texas Instruments, 805 F.2d at 1561. 141. Id. 142. Id. at 1567.


later time the accused products were designed. Hence, the structures used

in the imported calculators were not "equivalent" to the various structures

shown in the TI specification, t4~ The Federal Circuit agreed with the

administrative law judge that the recited functions were all carried out to

the letter in the accused products TM and also agreed that there was no

infr ingement . The court found that while the administrative law judge

may have held TI a bit too closely to its specification, ~4s the totality of the

changes in structure seen in the imported calculators were simply too

much to permit a finding of equivalency.t46

Texas Instruments was the beginning of a line of cases which

hammered home two concepts: (1) that for infringement, means-plus-

function expressions do not cover all means of performing the function;

and (2) the last clause of 112(6) can operate to preclude infr ingement even

when the recited functions are performed to the letter in the accused

product. The ne×~ decision in this line of cases came down the following

year and elaborated further on the role of 112(6) on this point.

B. 112(6) "Acts as a Restriction" in Infringement Analysis

In late 1987, a year after Texas Instruments, the Federal Circuit

decided Pennwalt Corp. v. Durand-Wayland, Inc. 147 The case involved

an apparatus for sorting fruit, by weight, color, or both. The Federal

Circuit affirmed a ruling of noninfr ingement , fin6ing neither literal

infr ingement nor infr ingement under the equitable doctrine of equiva-

lents.148 This en banc decision sought to refocus infr ingement analysis

under the equitable doctrine of equivalents, to base it on an element-by-

element approach instead of an invention-as-a-whole approach, t49 The

143. Id. at 1568. 144. The court stated: "TI correctly states, and the [administrative law judge] so found,

that every function described in the [TI] patent claims is performed by the accused calculators. ~ ld.

145. Id. (citing D.M.I., Inc. v. Deere & Co., 755 F.2d 1570 (Fed. Cir. 1985)). 146. ld. at 1568-70. 147. 833 F.2d 931 (Fed. Cir. 1987), cert. denied, 485 U.S. 961 (1988). 148. ld. at 939. 149. The groundwork for this shift had been laid in Perkin-Elmer Corp. v. Westinghouse

Elec. Corp., 822 F.2d 1528 (Fed. Cir. 1987). The en banc rehearing of Pennwalt was possibly motivated by a desire within the court to harmonize its panel decisions on the issue of whether the doctrine of equivalents should be applied on an element-by-element basis or an invention-as-a-whole basis.

On the issues involving the doctrine of equivalents, the element-by-element approach was favored by the majority, which ~:onsisted of Chief Judge Markey and Circuit Judges Bissell


case is instructive on several points relating to means-plus-function

limitations.

First the court took pains to distinguish 112(6) analyses from the

equitable doctr ine o f equivalents. 112(6) signifies that the literal scope of

a means-plus-funct ion expression is not the total class o f things which can

per form the function, but the subclass of things consisting of the

structures disclosed in the specification and their equivalents.~5° Whereas

112(6) applies only where the exact function recited in the claim is

per formed in the accused product, the equitable doctrine o f equivalents

may apply when the exact function recited is not pe r fo rmed by the

accused product, but a similar function is. 15j

The court thereupon aff i rmed the district j udge ' s rulings o f nonin-

f r ingement under 112(6) and under the doctrine of equivalents. On

112(6), the aff i rmance was based on the absence from the accused sorters

o f any kind of "posi t ion-indicat ing means" to locate a particular piece of

fruit. 152 Hence, 112(6) could not apply. On the doctrine o f equivalents,

the court found that no function similar to the position-indicating function

(author of the opinion), Friedman, Rich, Davis, Nies, and Archer, only three of whom (Rich, Nies, Archer) are in regular service on the court today. A vigorous dissent was filed by Senior Judge Bennett, and was joined by Senior Judge Cowen and Circuit Judges Smith and Newman. Of the dissenters only Judge Newman is in regular service on th°- court today. Judges Nies and Newman each filed a separate opinion in Pennwalt as well, stating their views favoring element-by-element comparison and invention-as-a-whole comparison, respectively.

The dissent did not challenge the majority regarding the interpretation of 112(6). It did, however, contend that the majority, in its application of law to the facts, had blurred the line between 112(6) and the equitable doctrine of equivalents. Pennwalt, 833 F.2d at 942-43.

150. Pennwalt, 833 F.2d at 933-34. Thus, under 112(6) a special kind of equivalency analysis is used to ascertain the literal scope of the expression. In the court's words:

Thus, section 112, paragraph 6, rules out the possibility that any and every means which performs the function specified in the claim literally satisfies that limitation. While encompassing equivalents of those disclosed in the specification, the provision, nevertheless, acts as a restriction on the literal satisfaction of a claim limitation.

Id. at 934. 151. The court, in distinguishing the two concepts, said:

If the required function is not performed exactly in the accused device, it must be borne in mind that section 112, paragraph 6, equivalency is not involved. Section 112, paragraph 6, plays no role in determining whether an equivalent function is performed by the accused device under the doctrine of equivalents.

/d. 152, Id. at 939.


was present in the accused sorters.t53 Hence, the equitable doctrine did

not apply either. There was simply no infringement.

After the significant elaboration of 112(6) in Pennwalt, a case was

needed in which the court could make a general statement on how the

restrictive language of 112(6) was to apply.

C. A Generalized "Proper Understanding" o f 112(6)

Johnston v. IVAC Corp. tS~ involved medical thermometers, each

consisting of a probe, a handle, and disposable covers for the probe. The

invention as claimed had to do with firmly affixing the cover to the probe

by use of a sharp edge on the periphery of the probe near where it joined

the handle. The 112(6) dispute centered on the phrase: "means for

inserting [the probe into thL cover and for] forcing said probe cover to

deform over said [probe] and causing said sharp edge to inscribe itself

fixedly into said probe cover. "~55

The district court gran 'ed summary judgment of noninfringement on

the ground that the insertion means in the accused structure did not

involve a sharp edge, but rather a retaining ring. The defendant's probes

were deliberately machined so as not to have any sharp edges, and relied

on a friction fit instead of a cutting fit. j56 However, the patentee's

evidence showed microscopic scratches on the inside of the defendant's

covers, and the patentee said this raised a triable issue as to the "sharp

edge" limitation.

The Federal Circuit held that the claim phrase "to inscribe itself" was,

properly interpreted, not literally met. In reaching that conclusion, the

court made the following important observation about means-plus-function

limitations in claims: "An element of a claim described as a means for

performing a function, if read literally, would encompass any means for

performing the function. But section 112 (6) operates to cut back on the

type o f means which could literally satisj3, the claim language." ~57

The Johnston court then amplified the role of 112(6): ' : "Proper ly

153. Id. 154. 885 F.2d 1574 (Fed. Cir. 1989). 155. Id. at 1578 (emphasis omitted). 156. Id. 157. ld. at 1580 (emphasis added) (citing Pennwalt and Data Line Corp. v. Mic~o

Technologies, Inc., 813 F.2d 1196, 1201 (Fed. Cir. 1987)). Neither Pennwalt norData Line used this "cutting back" language, but arguably that was the sense of both cases. Johnston is impor',ant for bringing out this meaning expressly.


understood section 112 para. 6 operates more like the reverse doctrine o f

equivalents than the doctrine o f equivalents because it restricts the scope

of the literal c la im language. "j58

This statement is the culminat ion of what had been developed in the

law o f means-plus-funct ion claims up to that time. 112(6) does in fact

work like the reverse doctrine o f equivalents, which essentially says that

broad c la im language will not operate to convey claim coverage over

structures far different f rom those shown in the patent 's disclosure

portion, t59 It "cuts back" on what would otherwise, by wording only,

appear to be covered by the claim. The cour t ' s observat ion suggests a

caveat to readers o f the Supreme Cour t ' s dictum in Graver Tank & Mfg.

Co. v. Linde Air Prods. Co. J6o that if there is literal reading of a claim on

an accused structure, there is infr ingement "and that is the end of it. "~6s

That statement of the Court was probably not meant to be taken universal-

ly, and it will not be so taken today.

The court in Johnston pointed out that 112(6) can never apply unless

the exact reci ted function is present in the accused structure.~6"- The court

concluded that the inscription-fit function recited in the patentee 's claims

was not present in the defendant ' s product, which used friction fitting

instead; hence, there was no infr ingement. ~63

The insightful v iew of 112(6) expressed in Johnston has been cited and

fol lowed in later cases, notably Jonsson v. Stanley Works I~ and Laitram

Corp. v. Rexnord, Inc. 165 Jonsson held that the scope of 112(6) equiva-

lents to which the specif icat ion 's structure was entitled was restricted by

virtue o f an argument presented to the PTO in order to secure al lowance

of the c la im over the prior art.166 Laitram, fol lowing as it did in the wake

158. Johnston, 885 F.2d at 1580. 159. See, e.g., Westinghouse Elec. Corp. v. Boyden Power-Brake Co., 170 U.S. 537

(1898). For a discussion of the doctrine of reverse equivalents in modern patent law, see 4 Donald S. Chisum, Patents (MB) § 18.0414] (Nov. 1994). Charles F. Pigott, Jr., Equivalents In Reverse, 48 J. PAT. OFF. SOC'Y. 291 (1966).

160. 339 U.S. 605 (1950). 161. Id. at 607. 162. Johnston, 885 F.2d at 1580. If the exact recited function is not present, the analysis

immediately turns to the equitable doctrine of equivalents, ld. 163. Id. at 1580-81. The court also found no equivalency under the doctrine of

equivalents, and hence affirmed the summary judgment of noninfringement. Id. at 1581. 164. 903 F.2d 812, 819 (Fed. Cir. 1990). 165. 939 F.2d 1533, 1536 (Fed. Cir. 1991). 166. Jonsson, 903 F.2d at 816. The patent applicant had argued that the word "diffuse,"

to describe light in a claim to a personnel-identification system for opening doors, meant light from multiple sources. The accused system used only one source, and thus was held noninfringing. Id.

186 ttarvard Journal of Law & Technology [Vol. 8

of the generalized understanding of 1 12(6) revealcd in Johnston, involved

a large number of ancillary 112(6) propositions which must now be

discussed in some detail.

D. The Ancillary Issues

Laitram shed light on several issues, none of which had been

expressly dealt with in earlier cases. First, the court held that a patentee

has the burden of proof in showing infringement of a 1 12(6) expression.

To meet this burden a patentee must show more than that the accused

infringing device performed the same function. The court also held that

the inclusion of some structural language with!n an otherwise functional

phrase in a claim cannot avoid the limiting impact of 112(6). Further-

more, the court made clear that the restrictive impact of 112(6) cannot be

avoided by interpreting the claim in light of other (presumably broader)

claims in the patent.

Laitram involved a conveyor-belt design that prevented the tipping of

articles on the belt. The district court had found infringement, but the

Federal Circuit reversed. The geometry of the individual links in the

belt, and particularly the way they were joined together, was at issue in

Laitram. The claim limitation in dispute recited a "means for joining"

the links together in a certain described way, which included a recitation

of such physical parameters as the ratio of length to width. The claim

clause in question recited that holes in certain of the link members were

coaxial with holes in other link members, and that sets of links had their

hole axes parallel to other sets' hole axes.

The district court had treated the recitation as structural rather than

functional and had therefore not applied the analysis of 112(6). The

Federal Circuit found this to be error. ~67 According to the court: "The

recitation of some structure in a means plus function element does not

preclude the applicability of section 112(6). For example, in this case,

the structural description in the joining means clause merely serves to

further specify the function of that means."t68

The suggestion here is that recital of some structure in a means-for

ctause--a common practice of many practitioners--will not avoid the

narrowing, or "cutting back," effect of 112(6).

167. Laitram, 939 F.2d at 1535-36. 168. ld. at 1536.

No. 1] Cris is in Patent Cove rage 187

The cour t next re i tera ted P e n n w a l t ' s concise explana t ion of how

1 12(6) works :

[Sect ion 112(6)] means exactly wha t it says: To de te rmine

whe the r a c la im l imi ta t ion is met l i terally, where expressed

as a means for p e r f o r m i n g a stated funct ion, the cour t must

compa re the accused s t ruc ture wi th the disclosed s t ructure ,

and must find equiva lent s t ructure as well as identi ty of

c la imed funct ion for that structure.t69

The cour t said it was an e r ro r of law for the distr ict j udge to have held

that a c l a im l imi ta t ion in means - fo r format was met "mere ly because there

was some means in the accused device that p e r f o r m e d that funct ion . . . .

The pa tentee mus t p rove , for l i teral in f r ingement , that the means in the

accused device is s t ruc tura l ly equiva lent to the means descr ibed in the

spec i f i ca t ion . " ~70

The cour t then wen t on to deal wi th an a r g u m e n t by the patentee based

on the doc t r ine o f " c l a im di f ferent ia t ion . ' ' m La i t ram, the patentee,

sought to pe r suade the cour t that inasmuch as the patent also conta ined a

c la im dependen t f rom the c la im-in-sui t , and the dependen t c la im reci ted

the s t ruc ture s h o w n in the specif icat ion, the independent c la im-in-sui t

mus t cove r s o m e t h i n g more . This t radi t ional c la im- in te rp re ta t ion

approach was re jec ted by the cour t on the g round that w h e n appl ied to a

means -p lus - func t ion c la im, the approach would run counte r to the

Congres s iona l c o m m a n d in 112(6) to cons t ra in the cons t ruc t ion o f

" m e a n s " express ions to the speci f ica t ion s t ruc tures and the i r equiva-

169. Id. (quoting from Pennwalt Corp. v. Durand-Wayland, Inc., 833 F.2d 9:~I, 934 (Fed. Cir. 1987), cert. denied, 485 U.S. 961 (1988)).

170. ld. 171. This doctrine of patent law says that in interpreting one claim of a patent it is often

helpful to look at a claim which ~depends from" the first claim. For example, a claim might read: "I. A process for treating biological material, comprising the steps of . . . . ~ Suppose there is in the patent a dependent claim, i.e.+ one that refers back to the first claim and is thus deemed by law to contain all of the first claim's words (35 U.S.C. § 112, para. 4 (1988)) plus some new features: "2. The process of claim 1, wherein said biological material is human biological material." According to the doctrine of claim differentiation, in interpreting the scope of claim I, a court looks at claim 2. The court notes that claim 2 narrows claim I by restricting the biological material to "human" biological material. This aids the judge in determining that claim 1 is not limited to human biological material, else the two claims would present a redundancy. Therefore, the court concludes that claim 1 embraces some non-human biological materials. For a further exposition of this doctrine and cases that have applied it, see 4 Donald S. Chisum, Patents (MB) § 18.0316] (Nov. 1994).


lents, tT:

Finally, in holding that the claim before it was not infringed under the

equitable doctrine of equivalents, the court gave some comfort to

patentees by commenting: "This inquiry of equivalency to the joining

means under the doctrine [of equivalentsl may not be as limited as under

section 112(6). ''173

The court seems to have left open the question of what happens after

!:12(6) is properly applied in claim interpretation. The suggestion is that

the doctrine may be available to a patentee in the right circumstances,

either to go outside the recited function or to go somehow outside the

range of specification equivalents afforded by 112(6). In either case, the

starting point for applying the equitable doctrine will have to be the cut-

back scope resulting from application of 112(6).

VI. FUNCTIONAL TERMINOLOGY OTHER

THAN "MEANS FOR" EXPRESSIONS

It is not known whether the jurisprudence now developed with respect

to "means for" expressions will apply to other, equally functional,

expressions. Many modern technical words define things by the functions

they perform. TM For example, in mechanics, a "drill" or a "screwdriv-

er"; in electronics, an "amplifier" or an "AND gate"; in chemistry, a

"reducing agent" or a "catalyst." Will these functional expressions

likewise fall under the limiting effect of 112(6)?

One case has indicated that the precise language "means for" need not

be present to trigger the narrowing effect of 112(6). Fairchild Semicon-

ductor Corp. v. Nintendo Co. 175 involved the claim language "locking

172. Laitram, 939 F.2d at 1538. 173. ld. at 1539. 174. Judge Lourie has suggested a contrary view, i.e.. that claim language normally

defines things by what they are, rather than what they do. In Int'l Visual Corp. v. Crown Metal Mfg. Co., 991 F.2d 768 (Fed. Cir. 1993), he said in a concurring opinion:

The tripartite tests often examine only what a device does, rather than what it is. Claims to machines, articles of manufacture, and compositions of matter generally define inventions by what they are, i.e., structurally. Thus, products may meet the tripartite tests in terms of what they do but be significantly different in terms of what they are.

ld. at 775. 175. 20 U.S.P.Q.2d (BNA) 1657 (W.D. Wash. 1994).


means having a detent, "~76 in connection with home video game equip-

ment. After voicing the proposition that means-plus-function expressions

describe what a thing does, whereas a structural element describes what

it " is , ''t77 the court went on to find that a "locking means" was functional

because it refers to an object that locks something, its The extension was

slight because at least the word "means" was present. One wonders

whether the result would have been any different if the claim had simply

said "lock." Such a term, like so many in science and technology, is

functional notwithstanding the absence of "means for" wording.

Policy-wise, there seems to be no reason to treat these realistically

functional terms any differently from the way we now treat claims

phrased as "means for" carrying out a specified function, namely, as

limited by the structures shown in the specification and drawings of the

patent. But, on the other hand, such a reading would effectively nullify

the concept of generic claiming, without which patents would be of far

less commercial worth to their owners.

Interestingly, it is the P T O - - s o long resistant to 112(6)--which seems

to be most forward-thinking about this 112(6) issue. The new guidelines

for examiners , z79 described above, delve into the point in some detail.

They list examples of claim language which would trigger 112(6),

including "a jet dr iving device so constructed and located on the rotor as

to drive the rotor. "~s° The guidelines note that the use of the word

"means" is unnecessary. ~s~ On the other hand, the guidelines say some

terms in claims, while using the word "means ," do not fall within 112(6).

For example, "plate means" or "wing means" are said to be non-

functional because neither defines any function to be performed p e r se . ~82

The guidelines also describe how to apply 112(6) to various result-

oriented method steps, is3 In fact, 112(6) might apply to limit the

176. A detent is typically a small ball that is fixed to one part and spring-biased into a recess or groove in another part, to prevent relative motion between the two parts until a force is applied sufficient to overcome the spring bias and push the ball out of the recess or groove. The ball thus "detains" the second part from moving.

177. Fairchild, 30 U.S.P.Q.2d (BNA) at 1657 ("A means plus function element is written to describe what an element of the invention does. A structural element refers to what the element is . ' ) .

178. Id. 179. See Van Horn, supra note 103.. 180. Id. at 571. 181. Id. 182. Id. 183. ld. A method or process is claimed by reciting a novel sequence of steps, e.g.,

mixing X and Y; heating the mixture to form a viscous mass; and extruding the mass into


seemingly broad reach of many step recitations. Suppose, for example,

that a claim recited: "heating said mixture to form a slurry," without

saying how this is done, where the specification says the heating should

be at 200 to 400 degrees Celsius for about three hours. Would someone

infringe who forms a slurry by heating in a very different range? Perhaps

some indication may appear in forthcoming cases regarding the extent to

which 112(6) can apply to method c la ims) 84 It may well turn out that

method claims are not infringed by competitors ' processes which appear

to meet the literal words of a result-oriented step in the claims, but which

are performed in a dissimilar way from that disclosed in the specification.

Such a result may be desirable, to move away from what some see as too

broad of scope for modern generic patent claims. However, it may prove

difficult to isolate which steps in a method claim are sufficiently broadly

recited to trigger the restriction of 112(6).

CONCLUSIONS

A. Would Walker Have Won Today ?

The above exploration has shown how the provisions of Section 112,

paragraph 6, arose in the 1952 Congressional response to the 1946

Halliburton v. Walker case, and that the modern cases have applied it

essentially as Mr. Walker wanted the prior case law to be applied by the

Supreme Court. Would it have changed the result in HaUiburton?

The question is a close one. The Supreme Court could today no

longer invalidate Walker ' s claim on the basis that means-plus-function

language is indefinite. But what would the claim's coverage be in

Walker ' s infringement action against Halliburton? Walker ' s claim t85

would first of all reach his specifically disclosed acoustic tuning p ipe- -a

structure Halliburton did not use. Pursuant to 112(6) it would also reach

"equivalents thereof," but it is unclear whether that includes Halliburton's

electrical filter. True, the filter was used for the same end result as the

acoustic tuning pipe: to accentuate echoes from the tubing collars so that

tiicv could be accurately counted in calculating how far down in the well

the surface of the oil was. However, it is debatable whether Halliburton's

a continuo~as band. 184. There is no doubt that the statute contemplates such applicability. See supra note 14. 185. For the claim language in issue, see supra note 35 and accompanying text.

No. I] Crisis in Patent Coverage 191

filter worked in "substantially the same way" as Walker ' s tuning pipe. ~s6

Walker ' s tuning pipe worked to accentuate the desired echoes by creating

standing sound waves of a certain frequency in the pipe. Halliburton's

filter, by contrast, took as its input the electrical analogs of all the sounds

coming from the well, and then electrically passed only those that had the

desired frequency. The "substantially the same way" issue could be

decided either for Halliburton or for Walker, depending on the court 's

perception of the degree of similarity of mode of operation.

B. Where is the Patent and Trademark Office Now Headed?

The PTO for a time did not agree with the Federal Circuit 's view that

112(6) constrains the effective scope of means-for claims during PTO

examination, either on statutory subject matter issues of Section 101 or on

prior art issues of Sections 102 and 103. The conflict was worked out on

the prior art front, but the PTO's position on statutory subject matter after

Alappat remains unknown.

It is predicted that the Commissioner will acquiesce to Alappat on

construing claims to determine statutory subject matter as he did to

Donaldson on construing claims to determine obviousness. Any other

result would lead to confusion in the examination process, which logically

must begin with an assessment of just what the patent applicant is seeking

by way of a patent grant. It would be anomalous to say that the same

claim has two different scopes, one for judging obviousness and one for

judging whether the claim embraces any nonstatutory subject matter.

C. Where is the Law Headed on Infringement?

The last clause of 112(6) has been uniformly applied in recent years

to constrict the literal coverage of means-for expressions in infringement

cases. 187 It is now established law that in infringement analysis the legal

186. See supra note 16 for the proposition that equivalency between two structures requires, inter alia, that they operate in substantially the same way.

187. Two recent nonprecedential Federal Circuit decisions have applied ll2(6)toconclude that infringement requires not only the function recited, but the use of structures equivalent to those shown in the specification. See Baltimore Therapeutic Equip. Co. v. Loredan Biomedical, Inc., 30 U.S.P.Q.2d (BNA) 1672, 1674, 1676-77 (Fed. Cir. 1994) (finding no infringement where claim recited "rotation registering means" and "brake means" plus their respective functions; and accused device operated differently from the structure disclosed in the patent); Rain Bird Sprinkler Mfg. Corp. v. Toro Co., No. 93-1493, 1994 WL 224167, at "I-2 (Fed. Cir. May 23, 1994) (affirming judgment of noninfringement where accused


reach of such expressions is limited to the specific structures shown in the

disclosure portion of the patent, plus the equivalents of those specific

structures. Is there a reason to treat other claim expressions differently?

There is not. Generic coverage is a commercial necessity in a

practical patent system, because virtually any narrowly worded claim may

be avoided by a competitor. However, the courts must realize that many

words commonly thought of as denoting "structure" are really just as

"functional" as a means-for expression, and should be treated the same

way in determining patent scope. Objects like screwdrivers, adder

circuits, and catalysts are known and identified by what they do. Patent

claims using such words should accordingly have the same reach as a

corresponding "means for" expression would have. In an age where

technical language seems more and more to be defined in functional terms

in science and industry, it seems idle to insist that a philosophical notion

of what things "are" should control the scope accorded to important

commercial rights. Recourse to the specifics of a patent's disclosure

seems the right approach in all cases.

device carried out recited function but did not employ structure equivalent to that shown in patent).

Date post:	08-Apr-2020
Category:	Documents
Upload:	others
View:	1 times
Download:	0 times

THE CRISIS IN PATENT COVERAGE: DEFINING SCOPE OF AN...

Documents